t5_dewata_reconstruct_task

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2374

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0005
• train_batch_size: 80
• eval_batch_size: 80
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 200
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9667	1.0	1077	0.8600
0.8636	2.0	2154	0.7781
0.8097	3.0	3231	0.7304
0.7762	4.0	4308	0.6996
0.7448	5.0	5385	0.6728
0.7234	6.0	6462	0.6502
0.7046	7.0	7539	0.6334
0.6907	8.0	8616	0.6204
0.6775	9.0	9693	0.6039
0.6622	10.0	10770	0.5915
0.6496	11.0	11847	0.5800
0.638	12.0	12924	0.5676
0.6275	13.0	14001	0.5576
0.6209	14.0	15078	0.5483
0.6115	15.0	16155	0.5389
0.6017	16.0	17232	0.5322
0.5937	17.0	18309	0.5215
0.5911	18.0	19386	0.5152
0.5842	19.0	20463	0.5086
0.573	20.0	21540	0.5021
0.5699	21.0	22617	0.4941
0.5653	22.0	23694	0.4886
0.5554	23.0	24771	0.4825
0.5476	24.0	25848	0.4782
0.5456	25.0	26925	0.4733
0.5427	26.0	28002	0.4680
0.5369	27.0	29079	0.4639
0.5345	28.0	30156	0.4588
0.5283	29.0	31233	0.4557
0.525	30.0	32310	0.4513
0.5204	31.0	33387	0.4469
0.5179	32.0	34464	0.4431
0.5156	33.0	35541	0.4419
0.5124	34.0	36618	0.4380
0.5101	35.0	37695	0.4339
0.5063	36.0	38772	0.4320
0.5041	37.0	39849	0.4301
0.5018	38.0	40926	0.4262
0.5015	39.0	42003	0.4250
0.4976	40.0	43080	0.4224
0.4954	41.0	44157	0.4200
0.4953	42.0	45234	0.4189
0.4899	43.0	46311	0.4179
0.4919	44.0	47388	0.4161
0.4905	45.0	48465	0.4149
0.4906	46.0	49542	0.4137
0.488	47.0	50619	0.4127
0.4862	48.0	51696	0.4123
0.4858	49.0	52773	0.4120
0.4854	50.0	53850	0.4119
0.4842	51.0	54927	0.4117
0.4873	52.0	56004	0.4107
0.4861	53.0	57081	0.4102
0.485	54.0	58158	0.4100
0.4845	55.0	59235	0.4098
0.4902	56.0	60312	0.4129
0.4891	57.0	61389	0.4089
0.4869	58.0	62466	0.4049
0.4816	59.0	63543	0.4035
0.4799	60.0	64620	0.3989
0.4788	61.0	65697	0.3958
0.4739	62.0	66774	0.3948
0.4715	63.0	67851	0.3897
0.4694	64.0	68928	0.3884
0.4647	65.0	70005	0.3852
0.4653	66.0	71082	0.3848
0.4615	67.0	72159	0.3810
0.4601	68.0	73236	0.3793
0.456	69.0	74313	0.3772
0.4562	70.0	75390	0.3761
0.4545	71.0	76467	0.3736
0.4496	72.0	77544	0.3697
0.4504	73.0	78621	0.3675
0.4481	74.0	79698	0.3662
0.4456	75.0	80775	0.3649
0.4432	76.0	81852	0.3632
0.4426	77.0	82929	0.3629
0.4398	78.0	84006	0.3589
0.4393	79.0	85083	0.3577
0.4372	80.0	86160	0.3566
0.4347	81.0	87237	0.3543
0.4345	82.0	88314	0.3534
0.4357	83.0	89391	0.3529
0.4323	84.0	90468	0.3510
0.4324	85.0	91545	0.3513
0.4305	86.0	92622	0.3500
0.4272	87.0	93699	0.3490
0.4286	88.0	94776	0.3468
0.4262	89.0	95853	0.3469
0.4282	90.0	96930	0.3466
0.4235	91.0	98007	0.3460
0.4251	92.0	99084	0.3446
0.4255	93.0	100161	0.3446
0.4259	94.0	101238	0.3431
0.4252	95.0	102315	0.3426
0.4223	96.0	103392	0.3427
0.4226	97.0	104469	0.3424
0.422	98.0	105546	0.3423
0.4214	99.0	106623	0.3421
0.4226	100.0	107700	0.3420
0.4283	101.0	108777	0.3452
0.4295	102.0	109854	0.3435
0.4289	103.0	110931	0.3413
0.4239	104.0	112008	0.3399
0.4205	105.0	113085	0.3377
0.4198	106.0	114162	0.3344
0.4163	107.0	115239	0.3338
0.4157	108.0	116316	0.3303
0.4119	109.0	117393	0.3275
0.4116	110.0	118470	0.3239
0.4085	111.0	119547	0.3215
0.4055	112.0	120624	0.3192
0.4056	113.0	121701	0.3184
0.4031	114.0	122778	0.3181
0.4	115.0	123855	0.3144
0.3978	116.0	124932	0.3120
0.3949	117.0	126009	0.3108
0.3941	118.0	127086	0.3074
0.3908	119.0	128163	0.3075
0.3888	120.0	129240	0.3038
0.3889	121.0	130317	0.3013
0.3875	122.0	131394	0.2988
0.3849	123.0	132471	0.2983
0.3821	124.0	133548	0.2960
0.3805	125.0	134625	0.2965
0.38	126.0	135702	0.2960
0.3789	127.0	136779	0.2923
0.3768	128.0	137856	0.2921
0.3749	129.0	138933	0.2891
0.3715	130.0	140010	0.2873
0.3716	131.0	141087	0.2862
0.367	132.0	142164	0.2852
0.3674	133.0	143241	0.2822
0.3667	134.0	144318	0.2825
0.3656	135.0	145395	0.2801
0.3645	136.0	146472	0.2777
0.3643	137.0	147549	0.2786
0.3615	138.0	148626	0.2750
0.36	139.0	149703	0.2747
0.3605	140.0	150780	0.2737
0.3558	141.0	151857	0.2726
0.357	142.0	152934	0.2697
0.3553	143.0	154011	0.2699
0.3557	144.0	155088	0.2693
0.3537	145.0	156165	0.2680
0.352	146.0	157242	0.2665
0.3499	147.0	158319	0.2666
0.3511	148.0	159396	0.2637
0.3483	149.0	160473	0.2636
0.3479	150.0	161550	0.2621
0.3466	151.0	162627	0.2600
0.3448	152.0	163704	0.2610
0.345	153.0	164781	0.2594
0.3439	154.0	165858	0.2595
0.3411	155.0	166935	0.2579
0.3414	156.0	168012	0.2583
0.3408	157.0	169089	0.2563
0.3393	158.0	170166	0.2550
0.34	159.0	171243	0.2559
0.3382	160.0	172320	0.2534
0.3379	161.0	173397	0.2520
0.3337	162.0	174474	0.2519
0.3356	163.0	175551	0.2523
0.333	164.0	176628	0.2502
0.3325	165.0	177705	0.2505
0.3326	166.0	178782	0.2497
0.3328	167.0	179859	0.2482
0.3305	168.0	180936	0.2495
0.3312	169.0	182013	0.2472
0.3302	170.0	183090	0.2471
0.3286	171.0	184167	0.2457
0.3288	172.0	185244	0.2456
0.3284	173.0	186321	0.2463
0.3285	174.0	187398	0.2440
0.327	175.0	188475	0.2434
0.3264	176.0	189552	0.2430
0.3264	177.0	190629	0.2435
0.3248	178.0	191706	0.2422
0.3233	179.0	192783	0.2419
0.324	180.0	193860	0.2420
0.3247	181.0	194937	0.2410
0.3235	182.0	196014	0.2405
0.3235	183.0	197091	0.2405
0.3219	184.0	198168	0.2402
0.3219	185.0	199245	0.2397
0.3212	186.0	200322	0.2404
0.3215	187.0	201399	0.2387
0.3203	188.0	202476	0.2393
0.3207	189.0	203553	0.2392
0.3205	190.0	204630	0.2387
0.3188	191.0	205707	0.2382
0.3182	192.0	206784	0.2382
0.3209	193.0	207861	0.2383
0.3199	194.0	208938	0.2379
0.3191	195.0	210015	0.2376
0.3174	196.0	211092	0.2377
0.3158	197.0	212169	0.2376
0.3188	198.0	213246	0.2378
0.3181	199.0	214323	0.2373
0.3181	200.0	215400	0.2374

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1